The ultimate goal of our renewal application is the discovery of genes that predispose to mental illnesses. While a number of genome-wide significant quantitative trait loci (QTL) have been localized for mental illnesses, these findings have yet to result in true gene identifications. Yet, progress in elucidating the pathophysiology o major mental disorders, and subsequent treatment interventions, is predicated on causal gene identification. In our renewal application, we will utilize exhaustive genomic information obtained from whole genome sequencing (WGS) to identify causal variants/genes influencing endophenotypes for schizophrenia, bipolar disorder and/or major depression. An endophenotype is a heritable trait that is genetically correlated with disease liability, providing greater power to localize disease-related genes than affection status alone. Rare variants appear to be important in mental illness. Pedigree-based studies represent an implicit enrichment strategy for identifying rare variants and a pedigree-specific rare functional variant can be sufficient to verify that a given gene is involved in phenotypic variation. In the initial phase of our study, we acquired neuroanatomic, neurophysiologic and neurocognitive endophenotypes for mental illness in 1350 Mexican Americans from randomly selected extended pedigrees. Using existing high density SNP data, we successfully localized multiple genome-wide significant QTLs influencing endophenotypic variation. We will now move beyond QTL localization to the identification of genes that influence these endophenotypes. Achieving this goal is greatly enhanced by the availability of WGS data on ~2100 individuals, including all endophenotyped subjects. Our specific aims are to: (1) acquire structural and functional brain images and conduct neuropsychological examinations on 600 additional Mexican American family members with WGS data but without brain-related endophenotypes; (2) identify causal variants underlying existing QTLs influencing mental illness-relevant endophenotypes; (3) perform agnostic pedigree-based genome-wide association using only functional non-synonymous coding variants or putative regulatory variants to identify additional genes/variants influencing brain endophenotypes; and (4) Test for pleiotropic effects of the most likely variants identified in Aims 2 & 3 in a sample of 1000 schizophrenia cases, 1000 bipolar depression cases, 1000 major depressive disorder cases and 1000 controls. Our collaborative project includes applications from John Blangero, Texas Biomedical Research Institute, and David C Glahn, Yale University. Subcontracts for phenotyping (UTHSCSA; RE Olvera) and image analysis (University of Maryland, P Kochunov) are also included. This renewal application is designed to extend our initial study by identifying the specific genes that influence mental illnes endophenotypes.